Neural-Network-Based Diagnosis Using 3-Dimensional Myocardial Architecture and Deformation: Demonstration for the Differentiation of Hypertrophic Cardiomyopathy

Abstract

The diagnosis of cardiomyopathy states may benefit from machine-learning (ML) based approaches, particularly to distinguish those states with similar phenotypic characteristics. Three-dimensional myocardial deformation analysis (3D-MDA) has been validated to provide standardized descriptors of myocardial architecture and deformation, and may therefore offer appropriate features for the training of ML-based diagnostic tools. We aimed to assess the feasibility of automated disease diagnosis using a neural network trained using 3D-MDA to discriminate hypertrophic cardiomyopathy (HCM) from its mimic states: cardiac amyloidosis (CA), Anderson–Fabry disease (AFD), and hypertensive cardiomyopathy (HTNcm). 3D-MDA data from 163 patients (mean age 53.1 ± 14.8 years; 68 females) with left ventricular hypertrophy (LVH) of known etiology was provided. Source imaging data was from cardiac magnetic resonance (CMR). Clinical diagnoses were as follows: 85 HCM, 30 HTNcm, 30 AFD, and 18 CA. A fully-connected-layer feed-forward neural was trained to distinguish HCM vs. other mimic states. Diagnostic performance was compared to threshold-based assessments of volumetric and strain-based CMR markers, in addition to baseline clinical patient characteristics. Threshold-based measures provided modest performance, the greatest area under the curve (AUC) being 0.70. Global strain parameters exhibited reduced performance, with AUC under 0.64. A neural network trained exclusively from 3D-MDA data achieved an AUC of 0.94 (sensitivity 0.92, specificity 0.90) when performing the same task. This study demonstrates that ML-based diagnosis of cardiomyopathy states performed exclusively from 3D-MDA is feasible and can distinguish HCM from mimic disease states. These findings suggest strong potential for computer-assisted diagnosis in clinical practice.

Machine Learning Approaches in Diagnosis, Prognosis and Treatment Selection of Cardiac Amyloidosis

“…Using automated disease identification to distinguish HCM from states that mirror it, such as CA, Anderson–Fabry disease, and hypertensive cardiomyopathy, was the goal of one study (HTNcm). To distinguish between HCM and other situations, a fully connected layer feed-forward neural network was developed [ 102 ]. This study revealed that it is possible to diagnose cardiomyopathy states using only 3D-MDA and that it is capable of differentiating between HCM and similar disease states.…”

Fabry Disease and the Heart: A Comprehensive Review

“…Cardiac magnetic resonance imaging (MRI) can detect cardiac involvement even when the LVH severity is mild, allowing to reclassify 21% of FD patients as having cardiac involvement that was previously unrecognized [ 38 ]. Of note, recent studies have suggested that machine learning applied to 3D myocardial architecture and deformation obtained by cardiac MRI may present increased ability to perform differential diagnosis of the cause of HCM [ 39 ].…”

Limits of Cardiovascular Adaptation During an Extreme Ultramarathon

“… 4 Commercial software (cvi42, Circle Cardiovascular Imaging) was used for all measurements except 3D-MDA, performed using previously validated software. 2 Volumes were indexed to height, given the anticipated change in weight during the event.…”

“…Such events present a unique opportunity to study the impact of extreme physiological demand on cardiac tissue health, chamber remodeling, contractile performance, and their resultant influence on cardiac hemodynamics. Contemporary cardiac magnetic resonance (CMR) imaging provides the unique capacity to study these in a single setting through the use of tissue mapping, 1 3-dimensional myocardial deformation analysis (3D-MDA), 2 and 4-dimensional flow analysis (4D-flow). 3 In this case report, we performed serial CMR evaluations in a 37-year-old athlete who ran 2.469 km to raise awareness for rare disease research.…”